New member
- Messages
- 7
I'm seeking guidance on training a GPT-3.5 model using educational documents I possess. I'm new to the process and unsure where to begin. I've heard about pre-processing data involving prompts, but I want the initial training to focus solely on the documents I provide before fine-tuning it with specific example prompts and responses. I attempted using Python for this purpose, but encountered issues like hitting token limits even after deleting files. Therefore, I'd like to start fresh. Can someone assist me in understanding this process?"